EPISODE · Dec 17, 2025 · 22 MIN
Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics
from Daily Paper Cast · host Jingwen Liang, Gengyu Wang
🤗 Upvotes: 34 | cs.LG Authors: Jingdi Lei, Di Zhang, Soujanya Poria Title: Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics Arxiv: http://arxiv.org/abs/2512.12602v1 Abstract: Linear-time attention and State Space Models (SSMs) promise to solve the quadratic cost bottleneck in long-context language models employing softmax attention. We introduce Error-Free Linear Attention (EFLA), a numerically stable, fully parallelism and generalized formulation of the delta rule. Specifically, we formulate the online learning update as a continuous-time dynamical system and prove that its exact solution is not only attainable but also computable in linear time with full parallelism. By leveraging the rank-1 structure of the dynamics matrix, we directly derive the exact closed-form solution effectively corresponding to the infinite-order Runge-Kutta method. This attention mechanism is theoretically free from error accumulation, perfectly capturing the continuous dynamics while preserving the linear-time complexity. Through an extensive suite of experiments, we show that EFLA enables robust performance in noisy environments, achieving lower language modeling perplexity and superior downstream benchmark performance than DeltaNet without introducing additional parameters. Our work provides a new theoretical foundation for building high-fidelity, scalable linear-time attention models.
What this episode covers
🤗 Upvotes: 34 | cs.LG Authors: Jingdi Lei, Di Zhang, Soujanya Poria Title: Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics Arxiv: http://arxiv.org/abs/2512.12602v1 Abstract: Linear-time attention and State Space Models (SSMs) promise to solve the quadratic cost bottleneck in long-context language models employing softmax attention. We introduce Error-Free Linear Attention (EFLA), a numerically stable, fully parallelism and generalized formulation of the delta rule. Specifically, we formulate the online learning update as a continuous-time dynamical system and prove that its exact solution is not only attainable but also computable in linear time with full parallelism. By leveraging the rank-1 structure of the dynamics matrix, we directly derive the exact closed-form solution effectively corresponding to the infinite-order Runge-Kutta method. This attention mechanism is theoretically free from error accumulation, perfectly capturing the continuous dynamics while preserving the linear-time complexity. Through an extensive suite of experiments, we show that EFLA enables robust performance in noisy environments, achieving lower language modeling perplexity and superior downstream benchmark performance than DeltaNet without introducing additional parameters. Our work provides a new theoretical foundation for building high-fidelity, scalable linear-time attention models.
NOW PLAYING
Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics
No transcript for this episode yet
Similar Episodes
No similar episodes found.